Inorganic Chemistry CUXTKIBUTIUS FROX THX I)EPARTIMEST OF CHEXISTRY,
UXIVERSITY OF WASHISGTON, SEATTLE, WASHINGTON
Mossbauer Spectra of Iron i n Na,[Fe(CO),] and Na [Fes(CO)llH1 and Comments Regarding the Structure of Fe3(CO)121 BY NILS E. ERICKSOS AND h. W. FXIRHXLL
Received February 8 , 1965 The chemical-isomeric shift of iron in the compound sa,[Fe(CO)a],in which iron is formally in a -2 oxidation state, is large and negative. A negative shift is associated with an increased electron density in the nucleus and in this case is probably due t o decreased shielding of the 3s electrons in the iron atom as the d electrons delocalize by T bonding. The Mossbauer patterns for both iYa[Fer(CO)ilH] and Fea(CO)lz show that in both compounds two of the iron atoms are equivalent while the third is different. The X-ray data and the Mossbauer data considered together indicate that both of these compounds have similar nonsymmetrical, bridged triangular structures.
hilossbauer spectra have been reported for iron in a variety of (formal) oxidation states.2 It has been found that the electron density in the nucleus, i\k(0)l2,is dependent on both the degree of covalency of the bonding and the formal oxidation state of the iron. Thus, in ionic (high-spin) iron compounds, chemical isomeric (CI) shift^,^ 6 , in the Mijssbauer spectra become more negative in the order F e f 2 , Fe+3, Fe+4, and Fe+6. (In Fe57, increasing negative values of 6 correspond to increased values of l\k(0)l2.) The increase in /\k(0)I2 with increasing positive oxidation state is due t o a decrease in electronic shielding of (primarily) the 3s electrons as 3d electrons are lost by i ~ n i z a t i o n . ~ Covalent (low-spin) compounds show the same general behavior. Thus, ferricyanide ion shows a more negative shift than does ferrocyanide, although the shifts of both of these ions are more negative than those found for ionic ferric compounds. The relatively large negative shifts associated with covalent compounds are thought to be due t o extensive d electron delocalization due t o 7r bonding.2,5,6 An alternative view7r8which appears to maintain that the u bond network dominates the shifts observed in iron compounds, presumably via forward donation of 4s electrons, is, ( 1 ) Taken in part from a thesis submitted by Xils E . Erickson in partial fulfillment of t h e requirements for t h e degree of Doctor of Philosophy, a t the University of Washington, Iuov. 1964. ( 2 ) E. Fluck. W. Kerler, and W. Seuwirth, Aizgew. Chem. l i z l e m . E d . Eizgl., 2, 277 (1968). somer shift,” and “isomeric shift.” (3) Also known a s “chemical shift, For this shift t o be seen in Mossbauer spectroscopy, there must he a change in both t h e excited and ground state nuclear charge radii and a difference in b€‘(0)lz between source and absorber nuclei. T h e change in nuclear charge radii is a fixed quantity for each particular isomeric transition and i t is only because there is a difference in IYr(O)’*due t o chemical bonding (neglecting sma!l temperature dependent shifts) t h a t one observes a n effect due t o t h e difference in nuclear charge radius. I t seems inappropriate therefore to refer t o this shift solely a s t h e “isomer” or “isomeric” shift. T h e term “chemical shift” would be more appropriate but might lead one t o erroneously conclude t h a t its interpretation was the same a s t h a t for chemical shifts in n.m.1.. spectroscopy. T h e primary difference is t h a t paired and unpaired electron density contributes inordinately strongly t o n.m.1.. shifts. We therefore propose t h a t Mossbauer shifts due t o differences in bonding be denoted a s chemical isomeric shifts reflecting their dependence on changes in both nuclear and chemical properties. (4) I,. R. Walker, G. K. U‘ertheim, and V.Jaccarino, Phys. Rev. Letters, 6, 98 (1961). ( 5 ) J. Danon, J . Chem. Phys., 39,236 (1963). ( 6 ) N. E. Erickson, P h . D . Thesis. University of Washington. 1964 (unpublished). (7) I
